[gmx-users] Re: gmx 4.6.1, Expanded ensemble: weird balancing factors

[Dejun Lin]

It seems to have something to do with the integrator/pressure-coupling.  When
I ran the tutorial on
http://www.alchemistry.org/wiki/GROMACS_4.6_example:_Ethanol_solvation_with_expanded_ensemble,
everything seems fine unless I switched to sd integrator instead of md-vv
(and Pcoupl from MTTK to berendsen), in which case SHAKE didn't seem to be
able to settle the constraints. Then I turned off the constraints and used a
0.1 fs timestep and until:

           Step           Time         Lambda
         505000       50.50000        0.00000

             MC-lambda information
  Wang-Landau incrementor is:   0.0023263
  N  CoulL   VdwL    Count   G(in kT)  dG(in kT)
  1  0.000  0.000       40    0.00000    0.12919
  2  0.200  0.000       58    0.12919   -0.02748 <<
  3  0.500  0.000       38    0.10171   -0.02334
  4  1.000  0.000       39    0.07837    0.02407
  5  1.000  0.200       48    0.10244    0.03248
  6  1.000  0.400       37    0.13492    0.03664
  7  1.000  0.600       35    0.17156   -0.09477
  8  1.000  0.800       54    0.07678    0.08130
  9  1.000  1.000       51    0.15809    0.00000

   Energies (kJ/mol)
           Bond          Angle Ryckaert-Bell.          LJ-14     Coulomb-14
    1.37839e+01    1.29774e+01    1.63229e+00   -4.16561e-01   -2.33737e+01
        LJ (SR)  Disper. corr.   Coulomb (SR)   Coul. recip.      Potential
    5.65457e+03   -2.31756e+02   -3.81717e+04   -3.22651e+03   -3.59708e+04
    Kinetic En.   Total Energy    Temperature Pres. DC (bar) Pressure (bar)
    6.59243e+03   -2.93784e+04    2.93987e+02   -1.41697e+02   -1.41675e+02
      dVcoul/dl       dVvdw/dl
   -1.40145e+01   -4.61130e+01


Step 505312  Warning: pressure scaling more than 1%, mu: 1.01715 1.01715
1.01715

Step 505313  Warning: pressure scaling more than 1%, mu: 1.01579 1.01579
1.01579

Step 505314  Warning: pressure scaling more than 1%, mu: 1.01503 1.01503
1.01503

Step 505315  Warning: pressure scaling more than 1%, mu: 1.01482 1.01482
1.01482

Step 505316  Warning: pressure scaling more than 1%, mu: 1.01368 1.01368
1.01368

Step 505317  Warning: pressure scaling more than 1%, mu: 1.01245 1.01245
1.01245

Step 505318  Warning: pressure scaling more than 1%, mu: 1.01138 1.01138
1.01138

Step 505319  Warning: pressure scaling more than 1%, mu: 1.01056 1.01056
1.01056

-------------------------------------------------------
Program mdrun_d, VERSION 4.6.3
Source code file:
/scratch/dlin13/gromacs/gromacs-4.6.3/gromacs-4.6.3/src/mdlib/domdec.c,
line: 4350

Fatal error:
A charge group moved too far between two domain decomposition steps
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
-------------------------------------------------------

the simulation just crashed. The simulation was fairly unstable if I tried
to restart from a checkpoint and I'm not comfortable to keep doing that. But
note that when the Wang-Landau incrementor was down to about 0.002, the dG
values were very close to the equilibrated values if I didn't modify the mdp
file from the tutorial. Any idea?




--
View this message in context: http://gromacs.5086.x6.nabble.com/gmx-4-6-1-Expanded-ensemble-weird-balancing-factors-tp5007681p5009942.html
Sent from the GROMACS Users Forum mailing list archive at Nabble.com.